Applying BigQuery javascript UDF to groups - google-bigquery

Question

I am trying to apply a javascript user defined function to groups. In the following code, --group by my_group within tuple is commented out. I want to apply the temp function test on every my_group within test_data. The code runs if group by is commented out. If I try to include group by, it produces a "scalar subquery produced more than one element". What change should I make so that I can output an array per group (my_group)?
#standardSQL create function
CREATE TEMP FUNCTION test(a ARRAY<STRING>)
RETURNS ARRAY< STRING >
LANGUAGE js AS '''
var combine = function(a) {
var fn = function(n, src, got, all) {
if (n == 0) {
if (got.length > 0) {
all[all.length] = got;
} return;
}
for (var j = 0; j < src.length; j++) {
fn(n - 1, src.slice(j + 1), got.concat([src[j]]), all);
} return;
}
var all = [];
for (var i = 1; i < a.length; i++) {
fn(i, a, [], all);
}
all.push(a);
return all;
}
return combine(a)
''';
WITH test_data AS (
SELECT 'Shirt' item, 'Cashless' my_group UNION ALL
SELECT 'Jeans', 'Cashless' UNION ALL
SELECT 'Jeans', 'Cash' UNION ALL
SELECT 'Cap', 'Cash' UNION ALL
SELECT 'Shirt', 'Cash' UNION ALL
SELECT 'Cap', 'Cashless'
),
tuple as (
SELECT ARRAY_AGG(DISTINCT item) items
FROM test_data
--group by my_group (uncommenting it creates error)
)
select * from unnest(test((select items from tuple)))
I am looking for an output like the following:
my_group Item
Cash Shirt
Cash Jeans
Cash Cap
Cash Shirt,Jeans
Cash Shirt,Cap
Cash Jeans,Cap
Cash Shirt,Jeans,Cap
Cashless Shirt
Cashless Jeans
Cashless Cap
Cashless Shirt,Jeans
Cashless Shirt,Cap
Cashless Jeans,Cap
Cashless Shirt,Jeans,Cap

Answer 1

Consider below (omitting the function piece to keep answer compact enough ...)
WITH test_data AS (
SELECT 'Shirt' item, 'Cashless' my_group UNION ALL
SELECT 'Jeans', 'Cashless' UNION ALL
SELECT 'Jeans', 'Cash' UNION ALL
SELECT 'Cap', 'Cash' UNION ALL
SELECT 'Shirt', 'Cash' UNION ALL
SELECT 'Cap', 'Cashless'
), tuple as (
SELECT my_group, ARRAY_AGG(DISTINCT item) items
FROM test_data
group by my_group
)
select my_group, item
from tuple,
unnest(test(items)) item
with output